iiinova — where your digital self keeps evolving

Ben Miller @bkmi.bsky.social · 14/05/2025

We've open sourced Adjoint Sampling! It's part of a bundled release showcasing FAIR's research and open source commitment to AI for science. github.com/facebookrese... There's also a blog about the release ai.meta.com/blog/meta-fa...

github.com

GitHub - facebookresearch/adjoint_sampling: code for "Adjoint Sampling: Highly Scalable Diffusion Samplers via Adjoint Matching"

code for "Adjoint Sampling: Highly Scalable Diffusion Samplers via Adjoint Matching" - facebookresearch/adjoint_sampling

0223

Ben Miller @bkmi.bsky.social · 14/05/2025

This was a significant community effort, so thank you to LBNL, Princeton, Genentech, Stanford, U. of Cambridge, CMU, NYU, LANL, the Roche Group and Berkeley for helping make this happen!

000

Ben Miller @bkmi.bsky.social · 14/05/2025

UMA, without fine-tuning to specific tasks, performs similarly or better in both accuracy and inference-speed/memory-efficiency than specialized models on a wide-range of material, molecular and catalysis benchmarks.

110

Ben Miller @bkmi.bsky.social · 14/05/2025

With all these large diverse datasets, we ask whether it’s possible to train a single model that performs well across several domains without the need for additional finetuning? Introducing UMA - our latest model trained on over 30 billion atoms across our open-science datasets.

100

Ben Miller @bkmi.bsky.social · 14/05/2025

Baseline models are released and evaluated on OOD evaluation tasks. Metrics that go beyond just a structure’s energy and forces are important for understanding how well models perform on downstream applications. We hope to release a public leaderboard in the near future!

100

Ben Miller @bkmi.bsky.social · 14/05/2025

With over 100M structures calculated at ωB97M-V/def2-TZVPD, with variable charge+spin, and spanning biomolecules, electrolytes, metal complexes, and community datasets, OMol25 represents the largest most diverse dataset of its kind.

100

Ben Miller @bkmi.bsky.social · 14/05/2025

While our earlier datasets (OC20, OC22, OMat24) focused on catalysts and materials, it was evident the community lacked a large, high quality, diverse molecular DFT dataset for training ML models. The Open Molecules 2025 (OMol25) Dataset hopes to accomplish just that.

100

Ben Miller @bkmi.bsky.social · 14/05/2025

Excited to share our latest releases to the FAIR Chemistry’s family of open datasets and models: OMol25 and UMA! @AIatMeta @OpenCatalyst OMol25: huggingface.co/facebook/OMo... UMA: huggingface.co/facebook/UMA Blog: ai.meta.com/blog/meta-fa... Demo: huggingface.co/spaces/faceb...

162

Ben Miller @bkmi.bsky.social · 03/05/2025

Really happy with this work led by Aaron! Take a look people! @franknoe.bsky.social @wellingmax.bsky.social

040

Reposted by Ben Miller

Chaitanya K. Joshi @chaitjo.bsky.social · 01/04/2025

Excited to be a keynote speaker at the Netherlands Society on Biomolecule Modelling annual meeting tomorrow! I'll be speaking about All Atom Diffusion Transformers for the first time publicly! Thank you @erikjbekkers.bsky.social for the invitation 🙏 www.nsbm.nl

031

Reposted by Ben Miller

ELLIS @ellis.eu · 19/03/2025

Meet @ivonajdenkoska.bsky.social, Postdoc Researcher at the Uni Amsterdam 🇳🇱. She researches multimodal #FoundationModels, developing efficient methods for vision-language understanding and generative tasks. Learn a career highlight of hers and her advice for young scientists. #WomenInELLIS #AI #ML

092

Reposted by Ben Miller

George Musser @georgemusser.com · 17/03/2025

God, that was hard to bear. Students told a physics town hall that grad schools have rescinded admission. NSF proposals are censored for words like “bias,” even when the meaning is technical. Physics curricula are being banned by school districts for encouraging girls. #APSGlobalSummit2025 #APSMarch

171182470

Ben Miller @bkmi.bsky.social · 11/03/2025

Chaitanya led a very impressive paper here! Congrats and yay for unified representations!

031

Reposted by Ben Miller

WOinActie.bsky.social @woinactie.bsky.social · 09/03/2025

Here is tomorrow’s program. We’re meeting at 11 a.m. at the Garenmarktplein in Leiden. The program starts at 11.30. The teach outs are throughout the day in The Hague and Leiden: www.universiteitleiden.nl/en/news/2025...

universiteitleiden.nl

10 maart: Universiteit Leiden start estafettestaking tegen bezuinigingen

Medewerkers van de Universiteit Leiden trappen maandag 10 maart de estafettestaking af van de Nederlandse universiteiten tegen de kabinetsbezuinigingen.

08234

Reposted by Ben Miller

Hank Green @hankgreen.bsky.social · 02/03/2025

We are experiencing an assault on science unparalleled by anything I’ve seen in my life. It’s not one issue or another anymore, the entire institution is under attack by the most powerful individuals in the country. This Friday, where will you be? standupforscience2025.org

standupforscience2025.org

STAND UP FOR SCIENCE

March 7, 2025. Washington DC and nationwide. Because science is for everyone.

6404400012016

Reposted by Ben Miller

ELLIS @ellis.eu · 11/02/2025

ELLIS welcomes the InvestAI initiative as a strong step toward strengthening AI research, computing infrastructure, and talent development in Europe. Read our full statement: ellis.eu/news/ellis-w... #AIActionSummit

ellis.eu

ELLIS Welcomes InvestAI Initiative as a Strong Endorsement of Its Mission

The ELLIS mission is to create a diverse European network that promotes research excellence and advances breakthroughs in AI, as well as a pan-European PhD program to educate the next generation of AI...

05611

Reposted by Ben Miller

Colette Delawalla, PhD @cdelawalla.bsky.social · 10/02/2025

Get in dorks, we're going protesting! STAND UP FOR SCIENCE WITH US ON MARCH 7TH, 2025 WASHINGTON DC AND EVERY STATE CAPITOL Because science is for everyone! Find us at standupforscience2025.org #standupforscience2025 #scienceforall #sciencenotsilence

11243139

Ben Miller @bkmi.bsky.social · 02/02/2025

Doesn't the timing of this not make sense?

000

Reposted by Ben Miller

Adrian Hill @adrianhill.de · 30/01/2025

You think Jacobian and Hessian matrices are prohibitively expensive to compute on your problem? Our latest preprint with @gdalle.bsky.social might change your mind! arxiv.org/abs/2501.17737 🧵1/8

Figure comparing automatic differentiation (AD) and automatic sparse differentiation (ASD).

(a) Given a function f, AD backends return a function computing vector-Jacobian products (VJPs). (b) Standard AD computes Jacobians row-by-row by evaluating VJPs with all standard basis vectors. (c) ASD reduces the number of VJP evaluations by first detecting a sparsity pattern of non-zero values, coloring orthogonal rows in the pattern and simultaneously evaluating VJPs of orthogonal rows. The concepts shown in this figure directly translate to forward-mode, which computes Jacobians column-by-column instead of row-by-row.

314028

Reposted by Ben Miller

Katie Mack @astrokatie.com · 28/01/2025

If you’re in the US and you’d like to know what projects and vital services federal grants currently fund in your state, you can search here: www.usaspending.gov And you can find the contact information for your elected representatives here: www.usa.gov/elected-offi... They need to hear from you.

usaspending.gov

USAspending.gov

14356303375

Reposted by Ben Miller

sbi - Simulation-based inference @sbi-devs.bsky.social · 28/01/2025

🙏 Please help us improve the SBI toolbox! 🙏 In preparation for the upcoming SBI Hackathon, we’re running a user study to learn what you like, what we can improve, and how we can grow. 👉 Please share your thoughts here: forms.gle/foHK7myV2oaK... Your input will make a big difference—thank you! 🙌

0188

Reposted by Ben Miller

David Zipper @davidzipper.bsky.social · 26/01/2025

USDOT has now deleted this webpage

30701269

Ben Miller @bkmi.bsky.social · 24/01/2025

Looks beautiful! I have heard it has some weird sharp edges still though... what do you think? Have you replaced latex entirely?

100

Reposted by Ben Miller

Larry Hunter @proflhunter.bsky.social · 23/01/2025

The NIH actions go far beyond the DEI crackdown. All external advisory meetings are canceled, in ones that were in progress today. That means that NIH cannot issue new grants, as external review is required by law. It's a big deal. See, e.g. www.science.org/content/arti...

science.org

Trump hits NIH with ‘devastating’ freezes on meetings, travel, communications, and hiring

Researchers facing

186

Ben Miller @bkmi.bsky.social · 20/01/2025

I sadly won't attend (at least not in person) this year, but it was great fun with great people!

141

Reposted by Ben Miller

Microsoft Research @msftresearch.bsky.social · 16/01/2025

Microsoft researchers introduce MatterGen, a model that can discover new materials tailored to specific needs—like efficient solar cells or CO2 recycling—advancing progress beyond trial-and-error experiments. www.microsoft.com/en-us/resear...

16226

Reposted by Ben Miller

Simon Olsson @smnlssn.bsky.social · 23/12/2024

Come join us on an exciting joint PhD project on ML for protein function with @kailalab.bsky.social -- We are looking for a strong quantitative candidate with prior experience in ML. The position is based in Gothenburg and is fully funded for 5 years. www.chalmers.se/en/about-cha...

chalmers.se

Vacancies

13016

Ben Miller @bkmi.bsky.social · 08/01/2025

This is the best post I've seen on bluesky so far. Impressive and very cool counterintuitive result with an extremely clear graphical explanation.

180

Reposted by Ben Miller

Brandon Amos @bdamos.bsky.social · 06/01/2025

📢 My team at Meta (including Yaron Lipman and Ricky Chen) is hiring a postdoctoral researcher to help us build the next generation of flow, transport, and diffusion models! Please apply here and message me: www.metacareers.com/jobs/1459691...

metacareers.com

Postdoctoral Researcher, Fundamental AI Research (PhD)

Meta's mission is to build the future of human connection and the technology that makes it possible.

15312

Ben Miller @bkmi.bsky.social · 26/12/2024

There is an opening to a book that clarifies this. I forgot which one... Indeed though, I find your terminology to be much better. The proposal should be identified specifically. All methods use the reject accept algorithm from metropolis hastings (green).

020

Reposted by Ben Miller

Joey Bose @joeybose.bsky.social · 18/12/2024

🔊 Super excited to announce the first ever Frontiers of Probabilistic Inference: Learning meets Sampling workshop at #ICLR2025 @iclr-conf.bsky.social! 🔗 website: sites.google.com/view/fpiwork... 🔥 Call for papers: sites.google.com/view/fpiwork... more details in thread below👇 🧵

28519

Ben Miller @bkmi.bsky.social · 19/12/2024

I think the paired episodes in the finale of season 2 are the best!

040

Reposted by Ben Miller

Bálint Máté @balintmate.bsky.social · 17/12/2024

hello bluesky! we have a new preprint on solvation free energies: tl;dr: We define an interpolating density by its sampling process, and learn the corresponding equilibrium potential with score matching. arxiv.org/abs/2410.15815 with @francois.fleuret.org and @tbereau.bsky.social (1/n)

13310

Ben Miller @bkmi.bsky.social · 18/12/2024

Hope you enjoyed California! :)

110

Reposted by Ben Miller

Grigory Bartosh @gbarto.bsky.social · 13/12/2024

🧑‍🏫Neural Flow Diffusion Models at #NeurIPS2024 tomorrow! Discover how to build learnable noising processes for straight-line generative trajectories end-to-end and without simulations!🤯 📍West Ballroom A-D #6809 ⏰Fri 13 Dec 4:30 pm — 7:30 pm 🔗https://neurips.cc/virtual/2024/poster/94656

0335

Ben Miller @bkmi.bsky.social · 09/12/2024

It is typical in academia that you are expected to move, perhaps every 2-4 years, until (and possibly after) you get a tenure track job. Sequential positions in the same place is widely seen as a sign of career narrowness. It may even be true. It seems the problem stems from this value system.

110

Ben Miller @bkmi.bsky.social · 08/12/2024

I'll be at NeurIPS. I'm still looking for an intern: Generative modeling for discrete & continuous variables that optimize reward functions, or doing model distillation on big MLIPs, or your cool idea! Contact me on whova or here!

1150

Ben Miller @bkmi.bsky.social · 07/12/2024

These outer and inner products remind me of the operations in a geometric / Clifford algebra. Can you comment on the connection?

100

Ben Miller @bkmi.bsky.social · 04/12/2024

Is there much literature on optimization using the Hellinger distance? Naively it seems challenging because it looks like one cannot Monte carlo sample the objective... however I guess the same issue exists for Wasserstein and it is heavily used!

100

Reposted by Ben Miller

Claudio Zeni @claudiozeni.bsky.social · 03/12/2024

🚨Our Machine Learning Force Field Mattersim is now available! 🚨 Check it out here 👇 msft.it/6013oBZLt The force field is designed to be used on a vast range of temperatures and pressures, try it yourself :) Feedback and suggestions are very welcome!

msft.it

GitHub - microsoft/mattersim: MatterSim: A deep learning atomistic model across elements, temperatures and pressures.

MatterSim: A deep learning atomistic model across elements, temperatures and pressures. - microsoft/mattersim

58125

Ben Miller @bkmi.bsky.social · 30/11/2024

Other questions: is it the alpha-stability of the gaussian that leads to gaussianity at infinity? That implies to me there should be other diffusion with other alpha-stable distributions, right?

100

Ben Miller @bkmi.bsky.social · 30/11/2024

Hello diffusion pros. I want to understand how one goes from Langevin dynamics to the fokker-planck equation. I mean in the statistical mechanics sense. Even better if this is generalized to other spaces (such as the torus). Anyone got a recommendation for me?

211